Apart from rs9804190, the externalizing- and PTSD-associated regions of ANK3 appeared to be distinct as, despite their close proximity, none of the PTSD-associated SNPs were nominally associated with externalizing or vice-versa. To assess whether these signals were, in fact, independent, we examined the LD structure across the regions. We found moderate-to weak LD across the region, with some pockets of long range LD within the region of PTSD-associated SNPs, indicated by clusters of high LD (red) within an otherwise low LD (white) region (Figure 2A). When we restricted our attention to the SNPs in Table 1 (Figure 2B), we found that the associated SNPs clearly fell into two clusters of LD, the externalizing-associated SNPs (SNPs 1 to 4), and the PTSD-associated SNPs and rs9804190 (SNPs 5 to 12). We then explored reasons for the observed long range LD within the region containing the PTSD-associated SNPs. When we estimated the haplotype frequency, it became clear that the PTSD-associated SNPs tag a single haplotype. Starting with the 3 SNPs in perfect LD, and adding adjacent SNPs that either increased the association or
